Carl June on CRISPR, CAR–T and how the Vietnam War dropped him into medicine
In August of 2011, Carl June and his team published a landmark paper showing their CAR–T treatment had cleared a patient of cancer. A year-to-the-month later, Jennifer Doudna made an even bigger splash when she published the first major CRISPR paper, setting off a decade of intense research and sometimes even more intense public debate over the ethics of what the gene-editing tool could do.
Last week, June, whose CAR–T work was eventually developed by Novartis into Kymriah, published in Science the first US paper showing how the two could be brought together. It was not only one of the first time scientists have combined the groundbreaking tools, but the first peer-reviewed American paper showing how CRISPR could be used in patients.
June used CRISPR to edit the cells of three patients with advanced blood cancer, deleting the traditional T cell receptor and then erasing the PD–1 gene, a move designed to “unleash” the immune cells. The therapy didn’t cure the patients, but the cells remained in the body for a median of 9 months, a major hurdle for the therapy.
Endpoints caught up with June about the long road both he and the field took to get here, if the treatment will ever scale up, and where CRISPR and other advancements can lead it.
The interview has been condensed and edited.
You’ve spoken in the past about how you started working in this field in the mid-90s after your wife passed away from cancer. What were some of those early efforts? How did you start?
Well, I graduated from high school and had a low draft number [for the Vietnam War] and was going to go to study engineering at Stanford, but I was drafted and went into the Naval Academy in 1971, and I did that so I wouldn’t have to go to the rice fields.
The war ended in ’73, ’74, so when I graduated in 1975, I was allowed to go to medical school, and then I had a long term commitment to the Navy because they paid for the Acadamy and Medical school. And I was interested in research and at the time, what the Navy cared about was a small scale nuclear disaster like in a submarine, and like what happened at Chernobyl and Fukushima. So they sent me to the Fred Hutchinson Cancer Center where I got trained in cancer, as a medical oncologist. I was going to open a bone marrow transplant center in Bethesda because the Navy wanted one in the event of a nuclear catastrophe.
And then in 1989, the Berlin Wall came down and there was no more Cold War. I had gone back to the Navy in ’86 for the transplant center, which never happened, so then I had to work in the lab full time. But in the Navy, all the research has to be about combat and casualty. They care about HIV, so my first papers were on malaria and infectious disease. And the first CAR-T trials were on HIV in the mid-90s.
In ’96, my wife got diagnosed with ovarian cancer and she was in remission for 3-4 years. I moved to the University of Pennsylvania in 1999 and started working on cancer because I wasn’t allowed to do that with the Navy. My wife was obviously a lot of motivation to do that. She passed away in 2001. Then I started working with David Porter on adoptive transfer T cells.
I got my first grant to do CAR-T cells on HIV in 2004, and I learned a whole lot. I was lucky to have worked on HIV because we did the first trials using lentiviruses, which is an engineered HIV virus.
I was trained in oncology, and then because of the Navy forced to work on HIV. It was actually a blessing in disguise.
So if you hadn’t been drafted, you would’ve become an engineer?
Yes. That’s what I was fully intending. My dad was a chemical engineer, my brother is an engineer. That’s what I thought I was going to do. No one in my family was ever a physician. It’s one of those many quirks of fate.
Back then, we didn’t have these aptitude tests. It was just haphazard. I applied to three schools — Berkeley, Stanford and Caltech — and I got into all three. It was just luck, fate.
And it turned out when I went to the Naval Academy, they had added a pre-med thing onto the curriculum the year before, so that’s what I did when I started, I did chemistry.
I would’ve [otherwise] been in nuclear submarines. The most interesting thing in the Navy then was the nuclear sub technology.
You talked about doing the first CAR-T trials on HIV patients because that’s where the funding was. Was it always in your head that this was eventually going to be something for cancer?
So I got out of the Navy in ’99 and moved to Penn. I started in ’98 working on treating leukemia, and then once I got to Penn, I continued working one day a week on HIV.
It’s kind of a Back-to-the-Future thing because now cancer has paved out a path to show that CAR–T cells can work and put down the manufacturing and it’s going to be a lot cheaper making it for HIV. I still think that’s going to happen.
Jim Riley, who used to be a postdoc in my lab, has some spectacular results in monkeys with HIV models. They have a large NIH and NIAID research program.
So we’re going to see more and more of that. The CAR technology is going to move outside of cancer, and into autoimmune and chronic infections.
I want to jump over to cytotoxic release syndrome (CRS) because a big part of the CRISPR study was that it didn’t provoke this potentially deadly adverse effect. When did you first become aware that CRS was going to be a problem?
I mean we saw it in the very first patient we treated but in all honesty, we missed it. I’m an MD, but I don’t see the patient and David Porter took care of the first three patients and our first pediatric patient, Emily Whitehead.
In our first patients, 2 out of 3, had complete remission and there were fevers and it was CRS but we thought it was just an infection, and we treated with antibiotics for 3 weeks and [eventually] it went away. And sort of miraculously he was in remission and is still in remission, 9 years later.
And then when we treated Emily. She was at a 106-degree fever over three days, and there was no infection.
I’ve told this story before. My daughter has rheumatoid arthritis, and I had been president of the Clinical Immunologists Society from 2009 to 2010, and the first good drug for juvenile rheumatoid arthritis that came out. I was invited to give the Japanese scientist Tadamitsu Kishimoto the presidential award for inventing the drug.
Then in 2012, Emily Whitehead was literally dying from CRS, she had multiple organ failures. And her labs came back and IL-6 levels were 1000x normal. It turns out the drug I was looking at for my daughter, it blocks IL-6 levels. I called the physician and I said, ‘listen there’s something actionable here, since it’s in your formulary to give it to her off-label.’
And she gave her the appropriate dose for rheumatoid arthritis. It was miraculous. She woke up very rapidly.
Now it’s co-labeled. When the FDA approved Kymriah, it was co-labeled. It kind of saved the field.
How were you feeling during this time? Did you have any idea what was happening to her?
No, not until we got the cytokine levels, and then it was really clear. The cytokine levels go up and it exactly coincided. Then we retroactively checked out adults and they had adverse reactions and it easy to see. We hadn’t been on the lookout because it wasn’t in our mouse models.
And it appeared with those who got cured. It’s one of the first on-target toxicities seen in cancer, a toxicity that happens when you get better. All the toxicities from chemotherapy are off-target: like leukopenia or hair loss.
I had a physician who had a fever of 106, I saw him on a fever when he was starting to get CRS. When the nurse came in and it said 106, they thought the thermometer must be broken. On Monday, I saw him, and said “how are you feeling” and he said “fine.” And I looked at the thermometer and his temperature was still 102.
People will willingly tolerate on-target toxicity — that’s very different from chemotherapy — if they know it helps get them better. That’s a new principle in cancer therapy.
You had these early CAR–T results almost at the same time that Doudna publishes the first CRISPR papers, then still in bacteria. When did you first start thinking about combining the two?
Yeah, it was published in Science in 2012 and that’s when Emily Whitehead got treated. It’s an amazing thing.
That’s something so orthogonal. You think ‘how in the heck can that ever benefit CAR–T cells?׳ but my lab had done the first edited cells in patients, published in 2012. And we used zinc-fingered nucleases, which were the predecessors to CRISPR. It knocked out one gene at a time, but we showed it was safe.
I was already into gene editing because it could make T cells resistant to HIV. So it was pretty obvious that there were candidates in T cells that you can knock out. And almost every lab started working on some with CRISPR, cause it was much easier.
We were the first to get full approval by the FDA, so we worked on it from 2012, had all the preclinical data by 2016, and then it takes a while to develop a lot of new assays for this as we were very cautious to optimize safety and it took longer than we wanted, but in the end, we learned a tremendous amount.
So what did we learn?
First of all our patients had advanced metastatic cancer and had had a lot of chemotherapy. The first patient had had 3 bone marrow transplants.
One thing is feasibility: could you really do all the complex engineering? So we found out we could. feasibility was passed.
Another was the fact that cas9 came out of bacteria, forms of strep and staph. Everyone has pre-existing immunity to Cas9 and we had experience from the first trial with Sangamo [with zinc-finger nucleases] where some patients had a very high fever. In that case, we had used adenoviruses, and it turned out our patients had very high levels of baseline immune response to adenoviruses, so we were worried that would happen with CRISPR, and it did not happen.
It did not have any toxicity. If it had, it would have really set the field back. If there was an immune response to cas9 and CRISPR, there could’ve been a real barrier to the field.
And then, the cells survived in the patients. The furthest on, it was 9 months. The cells had a very high level of survival. In the previous trials, the cells survived less than 7 days. In our case, the half-life was 85 days. We don’t know the mechanism yet.
And we found very big precision in the molecular scissors, and that was a good thing for the field. You could cut 3 different genes on 3 different chromosomes and have such high fidelity.
It [CRISPR] is living up to the hype. It’s going to fix all these diseases.
What’s the potential in CAR-T, specifically?
Well there’s many many genes that you can add. There are many genes that knocking out will make the cells work better. We started with the cell receptor. There are many, I think, academics and biotechs doing this now and it should make the cells more potent and less toxic.
And more broadly, what else are you looking at for the future of CAR–T? The week before your paper, there were the results from MD Anderson on natural killer cells.
Different cell types, natural killer cells, stem cells — putting CAR molecules into stem cells, macrophages. One of my graduate students started a company to do CAR macrophages and macrophages actually eat tumor cells, as opposed to T cells that punch holes in them.
There will be different cell types and there will be many more ways to edit cells. The prime editing and base editing. All different new variations.
You’ve talked about how people used to think the immuno-oncology, if it ever worked, would nevertheless be a boutique treatment. Despite all the advancements, Novartis and Gilead still have not met the sales they once hoped to grab from their CAR–T treatments. Are you confident CAR–T will ever be widely accessible?
Oh yeah, Novartis’ sales are going up. They had a hiccup launching.
Back in ’96 or ’97, when Genentech launched Herceptin, their commercial antibody, they couldn’t meet the demand either and then they scaled up and learned how to do better cultures. So right now Novartis is using tech invented in my lab in the 1990s culture tech that’s complex and requires a lot of labor, so the most expensive part is human labor. A lot can be made robotic. The scale problem will be much easier.
That’s an engineering problem that will become a thing of the past. The manufacturing problem will get a lot cheaper. Here in the US, we have a huge problem with how drugs are priced. We have a problem with pricing. That’s a political issue.
But in cell therapy, it’s just kind of the growth things you see in a new industry. It’ll get worked out.
This article has been updated to reflect that Jim Riley conducted work on CAR in HIV.